Recent online discussions about evidence-based medicine, artificial intelligence (AI), and Clinical Decision Support Systems (CDSS) revived my interest in Clinical Decision Support (CDS) and why it has been so slow to be adopted, and is so reviled by my physician colleagues. Obviously this is a huge subject, so I will focus on what I know best: the physician view.
As we have moved through the directives of the ACA and HITECH Act and implemented EHRs, complied with Meaningful Use, and moved on to MIPS and APMs, there has been a lot of talk about – and even a significant consulting industry built around — change management. The premise is that, if we just package and sequence the change correctly, the resistance will lie down and the new order will be adopted. Sounds more like marketing than operations.
I would argue that the neglected reality is change capacity. I am not speaking of organizational change capacity, but of individual change capacity; that aspect of all beings that only allows us to process a finite amount of alteration of our environment in a given period of time. We are, after all, impacting physicians one at a time under the umbrellas of organizations. The non-stop, top-down regulatory, reimbursement, workflow, reporting structure, and technological change onslaught that we have subjected our physicians to is overwhelming and has broken down nearly every physician that I speak to as I go around the country in my roles with the Kentucky Medical Association and the American Medical Association.
The main culprit is not change, in and of itself, but the lack of meaning and personalization in the change. Alerts that fire for every physician regardless of specialty or performance history aren’t seen as potential benefits to our patients, but as impediments to our work and delays in getting through our day and home to our loved ones. Even the best intentioned barrage of alerts or queries will wear down the alertee!
It reminds me of anecdote with my (then) 3-year-old grandson. He was riding home from preschool with his mommy, 18-month old brother, and doting aunt, who was alerting him to many of his favorite things and asking questions as they passed them. “Look at the firehouse. What color are firetrucks?” He dutifully answered, “Red.” Then, “There is a backhoe! You like construction equipment, right? What is your favorite type of construction equipment?” Compliantly he said, “Boom crane.” And then, “Did you see the little bunny rabbit? What do bunnies eat?” At which point his frazzled toddler mind snapped and he retorted, “I don’t want nobody asking me no more questions. I just need my peace and quiet!” To which my physician colleagues say, “Amen!”
There are ways to avoid the breakdowns and outbursts. I would suggest, starting at the individual level rather than organizational (we all throw around the terms Lean and Gemba in our C-Suites, but rarely live the meaning of those words), and working toward the following.
Prioritize the most effective and most needed alerts and associated CDS, as determined by the individual site based on its current quality environment, and initially present the top 8 to 10 alerts to all physicians. As an individual adheres to the recommended care at a preselected threshold, that alert drops off and the next item on the prioritization list begins to be presented. The same audit logic that recognized the individual’s adherence and achievement of threshold continues to monitor adherence; should the individual backslide, the alert returns until compliance again surpasses the threshold. In this way, the system automatically advances with the improved adherence of the clinicians, making room for the inevitable new alerts and practice advisories.
Additional alerts and associated CDS content specific to particular physician specialties will be identified and prioritized by each specialty through their service lines and applied in the manner outlined above, though likely applying fewer initial alerts.
I am well-aware of the shortcomings in our current CDS options and the EHRs we are trying to implement them within. However, there are better performers which can serve as starting points. CDS worthy of adoption will require a greater degree of sharing, both across physician groups and hospital systems and between EHR developers. There must be research to support the alerts and best practice advisories (BPAs) presented, and there must be rewards built into physician reimbursement and compensation systems, to make the inevitable time investment palatable. I also fully recognize that there are a multitude of vital care team members whose needs I have not addressed, but, as I stated at the start, I was going to address what I know best which is physicians. I think the personalized alert system I outlined can work for other care team members, but I will leave the determination of that to those best suited to comment.
The following is a good summary of some of additional things needed to support the previous recommendations:
- Development of CDS content that distills the wealth of information and clinical guidelines into a few action items that will have the biggest impact on patient-centered care.
- Learning from CDS implementing experience, including that related to incorporation into the EHR and delivery to the practitioner in a way that provides optimal support for the recommended clinical decisions.
- Practical strategies for embedding CDS in real-world environments that considers change management, people management, measurement of use, and usability considerations.
- Explication of the value proposition that fosters scale and spread of CDS through the development of clearinghouses and web-based repositories of CDS artifacts that can be shared, evaluated, and continuously improved through feedback from clinicians and patients (National Academy of Medicine White Paper).
I would recommend the above paper to anyone interested in CDS and advise you to not be put off by the 96-page PDF. Like many institutional publications, it is full of blank spaces, listings of participants, appendices, and summaries of preceding efforts. The real meat equates to about 30 pages of reading.
All of the above depends on overcoming hurdles that are beyond the scope of this piece and are being discussed extensively by others; interoperability, artificial and augmented intelligence, evidence-based clinical guidelines, and the myriad of other information technology issues that keep CIOs and CMIOs awake at night.
The immediate steps to overcoming these issues are respectful engagement of all stakeholders, encouraging their creative feedback, and beginning with the individual and managing and innovating up through organizations to regions and then national solutions.
[Currently working as a freelance consultant, David Bensema, MD, was previously both CIO and CMIO at Baptist Health Kentucky, where he led the implementation of an enterprise-wide EHR system. Bensema has served in various physician leadership roles and has a strong passion for advancing healthcare IT to improve patient care.]